Matches in SemOpenAlex for { <https://semopenalex.org/work/W4368342791> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4368342791 abstract "Processing a large number of proposals usually takes a significant proportion of inference time in two-stage object detection methods. Sparse regions with CNN features (Sparse R-CNN) was proposed using a small number of learnable proposals to replace the proposals derived from anchors. To decrease the missing rate, Sparse R-CNN uses six iterative detection heads to gradually regress the detection boxes to the corresponding objects, which hence increases the inference time. To reduce the number of iterative heads, we propose the iterative Hungarian assigner that encourages Sparse R-CNN to generate multiple proposals for each object at the inference stage. This decreases the missing rate when the number of iterative heads is small. As a result, Sparse R-CNN using the proposed assigner needs fewer iterative heads but gives higher detection accuracy. Also, we observe that the multi-layer outputs of the feature pyramid network contribute little to Sparse R-CNN and propose using a single-layer output neck to replace it. The single-layer output neck further improves the inference speed of Sparse R-CNN without the cost of detection accuracy. Experimental results show that the proposed iterative Hungarian assigner together with the single-layer output neck improves Sparse R-CNN by 2.5 AP50 on the Microsoft common objects in context (MS-COCO) dataset and improves Sparse R-CNN by 3.0 AP50 on the PASCAL visual object classes (VOC) dataset while decreasing 30% floating point operations (FLOPs)." @default.
- W4368342791 created "2023-05-05" @default.
- W4368342791 creator A5023214008 @default.
- W4368342791 creator A5057114272 @default.
- W4368342791 creator A5063328414 @default.
- W4368342791 date "2023-05-01" @default.
- W4368342791 modified "2023-10-18" @default.
- W4368342791 title "Compact Sparse R-CNN: Speeding up sparse R-CNN by reducing iterative detection heads and simplifying feature pyramid network" @default.
- W4368342791 cites W1536680647 @default.
- W4368342791 cites W1968245656 @default.
- W4368342791 cites W2031489346 @default.
- W4368342791 cites W2088049833 @default.
- W4368342791 cites W2109255472 @default.
- W4368342791 cites W3181008187 @default.
- W4368342791 cites W4281789597 @default.
- W4368342791 doi "https://doi.org/10.1063/5.0146453" @default.
- W4368342791 hasPublicationYear "2023" @default.
- W4368342791 type Work @default.
- W4368342791 citedByCount "0" @default.
- W4368342791 crossrefType "journal-article" @default.
- W4368342791 hasAuthorship W4368342791A5023214008 @default.
- W4368342791 hasAuthorship W4368342791A5057114272 @default.
- W4368342791 hasAuthorship W4368342791A5063328414 @default.
- W4368342791 hasBestOaLocation W43683427911 @default.
- W4368342791 hasConcept C11413529 @default.
- W4368342791 hasConcept C124066611 @default.
- W4368342791 hasConcept C138885662 @default.
- W4368342791 hasConcept C142575187 @default.
- W4368342791 hasConcept C153180895 @default.
- W4368342791 hasConcept C154945302 @default.
- W4368342791 hasConcept C159694833 @default.
- W4368342791 hasConcept C173608175 @default.
- W4368342791 hasConcept C199360897 @default.
- W4368342791 hasConcept C2524010 @default.
- W4368342791 hasConcept C2776151529 @default.
- W4368342791 hasConcept C2776214188 @default.
- W4368342791 hasConcept C2776401178 @default.
- W4368342791 hasConcept C33923547 @default.
- W4368342791 hasConcept C3826847 @default.
- W4368342791 hasConcept C41008148 @default.
- W4368342791 hasConcept C41895202 @default.
- W4368342791 hasConcept C75608658 @default.
- W4368342791 hasConcept C77637269 @default.
- W4368342791 hasConceptScore W4368342791C11413529 @default.
- W4368342791 hasConceptScore W4368342791C124066611 @default.
- W4368342791 hasConceptScore W4368342791C138885662 @default.
- W4368342791 hasConceptScore W4368342791C142575187 @default.
- W4368342791 hasConceptScore W4368342791C153180895 @default.
- W4368342791 hasConceptScore W4368342791C154945302 @default.
- W4368342791 hasConceptScore W4368342791C159694833 @default.
- W4368342791 hasConceptScore W4368342791C173608175 @default.
- W4368342791 hasConceptScore W4368342791C199360897 @default.
- W4368342791 hasConceptScore W4368342791C2524010 @default.
- W4368342791 hasConceptScore W4368342791C2776151529 @default.
- W4368342791 hasConceptScore W4368342791C2776214188 @default.
- W4368342791 hasConceptScore W4368342791C2776401178 @default.
- W4368342791 hasConceptScore W4368342791C33923547 @default.
- W4368342791 hasConceptScore W4368342791C3826847 @default.
- W4368342791 hasConceptScore W4368342791C41008148 @default.
- W4368342791 hasConceptScore W4368342791C41895202 @default.
- W4368342791 hasConceptScore W4368342791C75608658 @default.
- W4368342791 hasConceptScore W4368342791C77637269 @default.
- W4368342791 hasFunder F4320321001 @default.
- W4368342791 hasFunder F4320336363 @default.
- W4368342791 hasIssue "5" @default.
- W4368342791 hasLocation W43683427911 @default.
- W4368342791 hasOpenAccess W4368342791 @default.
- W4368342791 hasPrimaryLocation W43683427911 @default.
- W4368342791 hasRelatedWork W1931967218 @default.
- W4368342791 hasRelatedWork W2991276433 @default.
- W4368342791 hasRelatedWork W2993564273 @default.
- W4368342791 hasRelatedWork W3036924803 @default.
- W4368342791 hasRelatedWork W3116545149 @default.
- W4368342791 hasRelatedWork W3129447544 @default.
- W4368342791 hasRelatedWork W3177249605 @default.
- W4368342791 hasRelatedWork W3212509199 @default.
- W4368342791 hasRelatedWork W4285410089 @default.
- W4368342791 hasRelatedWork W4368342791 @default.
- W4368342791 hasVolume "13" @default.
- W4368342791 isParatext "false" @default.
- W4368342791 isRetracted "false" @default.
- W4368342791 workType "article" @default.